The Medicine That Does Not Ask You to Wake Up

A person with narcolepsy may be told to take a medicine before bed, then set an alarm for another dose in the middle of the night. It is an odd bargain: interrupt sleep to treat a disorder of sleep. That tension is where XWPharma found its opening. The company designed valiloxybate, an investigational oxybate medicine, for once-at-bedtime use. In September 2025, Avadel Pharmaceuticals agreed to pay $20 million upfront for broad rights to develop and sell it. The molecule still had clinical work ahead. The deal, however, had already put a price on a better-timed night.

That distinction between a completed deal and a completed medicine matters. Valiloxybate is a candidate, and the announced plans for additional clinical studies are plans. XWPharma has shown something narrower and still consequential: a specialist with a particular design problem can make an asset that a larger commercial organization wants to carry forward.

The clock as a design brief

XWPharma was founded in 2014 by Jia-Ning Xiang, a medicinal chemist whose earlier work included neuroscience research at GlaxoSmithKline and drug discovery at XenoPort. The company first appeared as XW Laboratories. Its home page describes a strategy of working with mechanisms that already have clinical support, then using chemistry and timing to address the limitations of existing therapies. This is a practical answer to a brutal problem in drug development: the brain offers many enticing theories and very few easy approvals.

Valiloxybate emerged from this approach. The candidate, formerly identified as XW10172 in company announcements, is a conjugate of oxybate. Its proposed use centers on sleep disorders including narcolepsy and idiopathic hypersomnia. XWPharma’s 2021 Phase 1 report described rapid oral absorption, release of oxybate and sustained concentrations suited to once-nightly dosing. Those are pharmacokinetic findings: they speak to how the substance moves through the body. They do not, by themselves, establish that the medicine improves a patient’s sleep or daily functioning.

The company had begun human testing in 2018. In 2019, the FDA granted orphan-drug designation for its narcolepsy candidate, then called XWL-008. The designation signaled a regulatory route for a rare condition, not an approval. By 2021, XWPharma was presenting Phase 1 data at the SLEEP annual meeting and discussing further trials in Parkinson’s-related sleep disorders and narcolepsy. The public record describes those ambitions; it should not be read as a list of completed studies.

“We are pleased to enter into this licensing agreement with Avadel, a recognized leader in the clinical and commercial advancement of once-at-bedtime oxybate therapy.”James Huang, XWPharma executive chairman, September 2025

What $20 million actually buys

The 2025 agreement gave Avadel exclusive rights to develop, manufacture and commercialize valiloxybate worldwide except mainland China, Hong Kong and Macau. The upfront payments totaled $20 million. XWPharma may receive up to $30 million in development milestones, additional sales milestones and tiered royalties if the product reaches commercial sales. A headline that simply adds every possible payment together would make an attractive number and a poor explanation. Most of the later money depends on future events.

The arrangement is a textbook example of XWPharma’s business model. It does discovery and early clinical development around patented CNS molecules. A partner with relevant development and commercial capacity can then take on broad rights, while XWPharma keeps contractual payments and potential royalties. The immediate customer in this transaction is a pharmaceutical company. Patients with narcolepsy or idiopathic hypersomnia would become users only if the medicine clears further testing and regulatory review.

Avadel brought a reason to listen. It already had LUMRYZ, a commercial once-at-bedtime oxybate product for narcolepsy. Valiloxybate was described in the joint announcement as salt-free and artificial sweetener-free, giving Avadel a candidate with a distinct formulation profile. In February 2026, Alkermes completed its acquisition of Avadel and said it intended to keep advancing valiloxybate in early clinical development. That is a transfer into a larger portfolio, not proof that the investigational medicine will reach pharmacy shelves.

A second bet, on the daytime brain

Sleep is only half of XWPharma’s public pipeline. XW10508 is an investigational oral, once-daily drug candidate aimed at major depressive disorder and chronic pain. The company describes it as a glutamatergic NMDA antagonist and AMPA activator, derived from esketamine chemistry and designed for extended release. It began a first-in-human study in 2021. In ordinary language, the pitch is to take a mechanism associated with fast-acting antidepressant effects and test whether a different chemical and dosing design can make it practical for regular oral use.

That is a hypothesis, not a patient benefit already on offer. Existing ketamine and esketamine therapies provide the comparison. XWPharma’s candidate would have to prove its own efficacy and safety; a familiar mechanism cannot do that work for it. The parallel with valiloxybate is revealing. In both programs, the company starts with biology that has reason to matter, then asks whether medicinal chemistry can change the experience of taking the drug.

The small company and the large system

The capital behind the work has been substantial. A 2017 Series B raised $17.5 million and brought cumulative announced financing to about $23 million. A $40 million Series C followed in 2020, led by Panacea Venture, with returning support from Johnson & Johnson Innovation and WI Harper Group among others. The same year XWPharma adopted its current name and appointed Leonard Blum chief executive. Those are disclosed financing amounts, not a running estimate of cash still available today.

The geography is equally telling. The company lists an address in Redwood City and operations in Wuhan and Taiwan. Its research, investors and licensing counterparty span more than one market. For a small drug developer, that reach is less about selling everywhere from day one than about assembling the science, capital, trials and commercial partner that a medicine may need over many years.

There is a useful lesson here for anyone building a specialized business. XWPharma did not ask a partner to believe in a vague promise of better neuroscience. It gave one drug a concrete design brief, generated early human data relevant to that brief and sold rights under a contract that assigns future work and future upside. This pattern depends on a known unmet need, defensible intellectual property, credible measurements and a buyer equipped to finish the job. Remove any one of those and the licensing logic becomes much weaker.

The most human detail is still the alarm clock. A drug meant to restore sleep should, if possible, leave sleep alone. XWPharma turned that observation into a molecule and then into a deal. Whether the molecule becomes a medicine remains a clinical question. The business result is already legible: the right answer to a patient’s inconvenience can be valuable long before it is available by prescription.